In a logistic system, it is often required to deliver different kinds of articles from one location to another, for example delivering goods from a warehouse to a sales site. In the case that a delivery task needs to be completed at multiple times, computer technology may be utilized to generate a loading plan to indicate which articles as well as how many articles are to be delivered during each transportation, so as to implement the delivery task within minimal times, thereby leveraging the transportation capability more effectively. For example, a transportation tool and articles may be modeled such that the transportation tool and various kinds of articles are modeled to have a solid shape with a certain geometric size, and then simulation technology may be used to acquire a loading plan by which solid shapes representing articles may be contained in the solid shape representing the transportation tool as many as possible is calculated with a simulation technology.
However, generating a shipping plan with the modeling method requires consumption of large amount of storage and computing resources. It should be easily understood that it is required at least to store data describing the model, and considerable operations are necessary for performing a three-dimensional simulation. Due to the irregularity of the solid shapes of articles, if the established model is too simple, space of the transportation tool must be wasted; however, on the other hand, if the established model is too complex, the data required to describe the model and the complexity for three-dimensional simulation operations will be increased. Besides, the modeling method cannot involve other factors in actual condition that affect loading other than solid shapes of the transportation tool and articles, for example, the articles' compressibility, bendability, and fragility, etc.